Data Methodology
1. Institutional Grade Data
Our platform is powered by the Financial Modeling Prep (FMP) API, providing us with direct access to SEC filings. Unlike standard retail platforms that cap data at 5 years, our engine processes up to 30 years of historical financial statements.
2. The Quant Engine
We do not rely on pre-calculated ratios that fail to account for historical stock splits. Instead, we pull raw /stable/ endpoints (Income Statement, Balance Sheet, Cash Flow) and calculate metrics strictly using Trailing Twelve Months (TTM) and current Enterprise Value to ensure perfect historical accuracy.
3. AI Quality Control
All narrative analyses are generated using a strictly constrained programmatic pipeline. To prevent AI hallucinations, mathematical data is hardcoded via our Next.js frontend using Tremor charts, while the text synthesis undergoes a strict Human-in-the-Loop review process before publication.